Current required for tube/valve amp bias circuit...

[yankeerob]

Hello all,

I have a couple of left over mains power transformers that were originally designed to go in a cathode biased amp - hence they do not have a bias tap in the HT secondary. I would like to purchase something like a small 0-115, 0-115 primary/0-24, 0-24 secondary transformer to supply a bias circuit for a fixed bias design. Question is... how much current is required?

The HT winding is 0-340V@200mA - these amps generally settle down at around 475V plates with a pair of 6L6/5881s...

Many thanks!!

[stokes]

The bias supply requires very insignificant current.No need to use a seperate tranny for your bias supply,you can just tap one of your high voltage secondaries to feed the bias supply.I'd suggest using a 220k/2 watt resistor from one leg of the "red" leads from the HV before your rectifier.From this 220k just add your bias circuit.It is a very common way to supply the bias circuit.

[yankeerob]

The bias supply requires very insignificant current.No need to use a seperate tranny for your bias supply,you can just tap one of your high voltage secondaries to feed the bias supply.I'd suggest using a 220k/2 watt resistor from one leg of the "red" leads from the HV before your rectifier.From this 220k just add your bias circuit.It is a very common way to supply the bias circuit.

Hey buddy, thanks for that... I thought it was low current but couldn't find where I'd read it

That said - my preferred method of standby switching is a DPST switch betwen the HT winding and the rectifier - particularly on tube/valve rectified power supplies. If I tap off the HT winding side of the switch with the resistor as described will this give me constant current to the bias supply circuit or will it be dependent on current flowing to the rectifier? I hope I'm making sense... I've had it before where there was no bias supply until the HT winding was engaged with the rectifier but seem to recall there being a cap and resistor network before the diode in those instances... I wondered what effect - if any - this may have had on tube/valve life...

Many sincere thanks!!

[Tom Phillips]

Some thoughts for your consideration.

...my preferred method of standby switching is a DPST switch betwen the HT winding and the rectifier - particularly on tube/valve rectified power supplies...

That is going to cause extra stress on the tube rectifier because, when you switch to operate, the rectifier will be fully warmed up and the first filter cap will be completely discharged. These conditions maximize the turn on current surge.

EDIT: I got off on the wrong path after that mention of "tube/valve rectified power supplies" and spaced out on the non tapped HV secondary of the transformer being used. The corrections are covered by the other members further down in this thread.
Cheers,
Tom

...If I tap off the HT winding side of the switch with the resistor as described will this give me constant current to the bias supply circuit or will it be dependent on current flowing to the rectifier? ...

You will indeed have constant bias voltage which is what you want. Actually the bias voltage will change a little when the plate supply is engaged but that’s not a significant concern. You are on the right track to design so that the bias supply is always on.

...I hope I'm making sense... I've had it before where there was no bias supply until the HT winding was engaged with the rectifier but seem to recall there being a cap and resistor network before the diode in those instances...

A capacitor coupled bias supply is required when the bias supply is derived from a full wave bridge rectifier. Your tube rectifier will be a full wave center tapped circuit so all will work fine.


You can see some examples of the bias supply tapped off the HV by looking at vintage Marshall Schematics.

[J M Fahey]

The HT winding is 0-340V@200mA

There's just one HV winding, so neither

high voltage secondaries to feed the bias supply.I'd suggest using a 220k/2 watt resistor from one leg of the "red" leads from the HV before your rectifier.From this 220k just add your bias circuit.

nor

the rectifier will be fully warmed up

Your tube rectifier will be a full wave center tapped circuit so all will work fine.
You can see some examples of the bias supply tapped off the HV by looking at vintage Marshall Schematics.

because with a single HV winding all you can use an SS bridge.

This will work:

A capacitor coupled bias supply is required when the bias supply is derived from a full wave bridge rectifier.

so look at a more modern Marshall such as a JCM900

[yankeerob]

Some thoughts for your consideration.


That is going to cause extra stress on the tube rectifier because, when you switch to operate, the rectifier will be fully warmed up and the first filter cap will be completely discharged. These conditions maximize the turn on current surge.

I'm pretty anal about big f-off current limiting resistors on the HT inputs of the tube recto on amps with those employed for that very reason and also use the minimum value 1st stage reservoir cap with a 220k/2W bleed resistor... so far so good on that and - in the main - only build them to order employing NOS rectos... Not being a smartarse... they've been OK so far...

You will indeed have constant bias voltage which is what you want. Actually the bias voltage will change a little when the plate supply is engaged but that’s not a significant concern. You are on the right track to design so that the bias supply is always on. A capacitor coupled bias supply is required when the bias supply is derived from a full wave bridge rectifier. Your tube rectifier will be a full wave center tapped circuit so all will work fine.

Right... now we're getting down to the nitty gritty... if I'm employing a FWBR (ie non-CTed HT) then a capacitor coupled bias supply IS required? If that is the case then an independent transformer providing -60V+ after rectification is a far better, predictable and therefore economical solution. If 0-24, 0-24 @10VA in series will get the job done then it's a no brainer... I'll share the schematic so that everyone can benefit... These power trannies were a low cost/reduced weight solution and I just need to move them on and start fresh.

You can see some examples of the bias supply tapped off the HV by looking at vintage Marshall Schematics.

Thanks Tom... yer a star!!

[yankeerob]

Thanks to everyone who has contributed to this thread - it's pretty much answered my question.

I can have an 'always ON' bias supply for less than 6 bucks (£3.85 GBP) using a small secondary transformer to provide me with 0-48V before rectification and 12VA (200mA+) is going the bury the job... probably seems like overkill but they're tiny and the least expensive option (lower VA ones are more expensive believe it or not!!!). I knew there was a reason we had to use the capacitor coupled design and I really don't like the idea of the output tubes having to wait for the current to flow and that type of bias supply eventually providing the voltage when it's good and ready - sometimes several seconds later... it can't be as good as an always on bias voltage... in my thoughts anyway... 'sides - there are only two sets to get rid of

Sincere thanks again to you all!!... when I've got 'em going I'll post some pics...

[g1]

If you want to use a transformer, that is fine. But as to it being more "economical", the cap coupled bias circuit mentioned is much cheaper.
As mentioned by JM, you can see them in JCM900 Marshall's, also in V series Ampeg's like the V4.
See attached, C16 for the marshall (from lower end of standby switch), C15 for the Ampeg.





P.S. Not sure what you meant about the "always on" bias, that does not have to do with the cap coupled method, it has to do with which side of the standby switch it is derived from in the other method.

[yankeerob]

Thanks for your input too g-one...

I would just dearly love to a.) take the guess-work out of the equation and b.) have the bias voltage already there when the B+ makes it's way to the plates... if the reality of the situation is that the B+ and bias supply charge up at pretty much the same pace and there is nothing to get worked up about in any case then I'll put a cork in my cake hole and get on with it

Can anyone explain how to calculate the value cap and resistor required? Many thanks for all the input!!!

[g1]

Just to be clear, you are using a solid-state bridge rectifier, like in the ampeg circuit above? Just with lower B+, around 475V ?

[yankeerob]

Yes g-one standard FWBR...

[g1]

The 5881 version of the 4500 Marshall has around 475V B+ so you should be able to use those same values. For C12 if you don't have classX, use at least 600V or better yet 1000V. It gets connected to the 340V winding of your PT.

[yankeerob]

That looks like a winner - I'm pretty certain I have some 630V rated 47nFs lying around... if not Farnell is never far away and they list a class X 47nF/1KV for pennies...

Is the purpose of the couling cap to provide isolation from the main rectification circuit?? I'm just trying to get my head around how one would calculate the cap value... the resistors have obvious purpose as do the electrolytics like any other bias supply circuit...

Many thanks again!!

[pdf64]

That's a primary function, but with this particular design, its capacitive reactance at 50Hz is being used with the 56k load resistor to create a frequency dependant potential divider.
A minor downside of this approach is that the bias supply voltage will be slightly higher at 60Hz than 50Hz.
Alternatively, a much higher value cap could have been used to provide isolation, but then a series dropper resistor would be required to bring the voltage down to a suitable level. And a higher value cap would likely be bulkier and more expensive.

[yankeerob]

That's a primary function, but with this particular design, its capacitive reactance at 50Hz is being used with the 56k load resistor to create a frequency dependant potential divider.
A minor downside of this approach is that the bias supply voltage will be slightly higher at 60Hz than 50Hz.
Alternatively, a much higher value cap could have been used to provide isolation, but then a series dropper resistor would be required to bring the voltage down to a suitable level. And a higher value cap would likely be bulkier and more expensive.

Re the bigger cap and series resistor/frequency dependent shananigans... this is where 4 quid means I haven't got to go messing around now/and supporting the amp unnecessarily in the future - for that amount I can get a relatively small, lightweight dual primary (0-115 x 2) dual secondary (0-24V x 2) 10VA transfomer that I can wire in parallel with the the mains tranny and follow my usual conventions - which simply involves tying the 0V side to ground, the 48V side through the diode (supported by a 100uF/100V ELC) and paralleled into 2 x 10K resistors the ends of each feeding the inputs of two 47K trim pots - the wipers (supported by 22uF/100V ELCs) feed independent bias feed resistors with the range being fine tuned with fixed value resistors in series to ground at the other end of the pots... allowing me to bias each side of the output independently. Over the top?... perhaps. Worth it? IMHO - definitely. With a really well matched pair/quad of valves the results are sublime... and I can usually get enough range to allow for any 6L6/KT66/5881/KT77 type... even 6550s at a push in a 2204 type design...

Many huge thanks to everyone who has contributed - even if you think I'm being stubborn... pdf64 has highlighted a definite advantage in spending a couple of quid (as opposed to a few pence) more to get what is likely to be a less awkward result long term... Now - to decide head or combo...